General pharmacology and pregnancy

Placental transfer of drugs has been primarily studied in sheep and laboratory animals, and direct extrapolation to dogs may be misleading due to species differences in placentation, extent of placental metabolism and transport of drug across the placenta 4. However, in general, the same physiochemical properties that allow drugs to cross the blood brain barrier also facilitate their placental transfer. The safest assumption is that most, if not all, drugs cross the placenta and affect the fetus. Elective procedures requiring anesthesia are best avoided during the first trimester (20 days of gestation in dogs), when the fetuses are most vulnerable to teratogenic effects of drugs.

Simple diffusion is the most important mechanism of placental transfer of drugs. Properties favoring transfer include: 

• Molecular weight < 600 Da
• High lipid solubility
• Low degree of protein binding
• Non-ionization at maternal blood pH 3 4

Most anesthetic drugs, with the exception of glycopyrrolate and neuromuscular blocking agents, have MW < 300 Da and are relatively lipid soluble, and thus readily cross the placenta. 

Drug protein binding and degree of ionization is determined by its pKa and blood pH, which in turn can affect the distribution between the dam and fetus. As blood pH decreases, acidic drugs such as thiobarbiturates will be less ionized and the fraction of drug bound to protein is decreased, leading to greater clinical effect 3 4. Weakly basic drugs (opioids, local anesthetics) become more highly ionized, resulting in less effect on the dam and fetus 3 4. Redistribution of drugs out of the fetus back into the dam’s circulation as maternal plasma levels decrease makes clinical estimation of fetal plasma concentrations difficult. Although ~50% of umbilical vein blood passes through the fetal liver, microsomal enzyme activity and metabolism are minimal 4.

Inhalant agents readily cross the placenta and should be titrated to the lowest level needed to achieve adequate anesthesia. Unlike halothane and methoxyflurane, which rely heavily on metabolism (~20-50% and 50-75% respectively) for elimination, isoflurane and sevoflurane are almost entirely eliminated through the respiratory system. Additionally, the low blood solubility results in rapid clearance from the newborn as long as they are breathing at delivery. It is important to avoid high inhalant concentrations to prevent neonatal respiratory depression and apnea. Isoflurane is associated with improved puppy survival at 7 days compared to methoxyflurane and there is no difference when compared to epidural anesthesia 1. 

• Anticholinergic agents such as atropine and glycopyrrolate are primarily used to decrease vagal tone due to opioids or traction on the uterus, or to support fetal heart rate. Choice depends on the desire for placental transfer, since atropine crosses the placental barrier but glycopyrrolate does not. Glycopyrrolate will mitigate the increased vagal tone caused by mu agonist opioids and prevent maternal bradycardia and possible hypotension. It also increases gastric pH and may decrease the severity of chemical pneumonitis should regurgitation and aspiration occur in the dam 3. Fetal bradycardia (< 150 beats/min) indicates fetal distress and is one of the prime indicators for emergency CS 10. Fetal cardiac output is more dependent on heart rate than blood pressure. Atropine may be administered to the dam if the objective is to increase fetal heart rate, which may be low due to hypoxemia, fetal distress or mu agonist opioids. Increasing the fetal heart rate will increase myocardial oxygen consumption in the face of hypoxemia, which may lead to myocardial ischemia, so atropine use is controversial. However, it may support the heart rate long enough to allow delivery and proper resuscitation. It is important to optimize maternal oxygenation, cardiac output and blood pressure, and ensure good ventilation for the puppies after delivery.
  
  

• Tranquilizers/sedatives are typically avoided due to cardiorespiratory and CNS depression. Xylazine alone or in combination with ketamine is associated with higher fetal death 1. Medetomidine at low doses (< 20 µg/kg) does not increase uterine muscle activity or cause abortion 3, but both medetomidine and dexmedetomidine cause significant decreased cardiac output in the dam due to vasoconstriction and baroreceptor-mediated bradycardia, and drug label recommendations advise against their use in pregnant dogs. Benzodiazepines (diazepam and midazolam) can cause neonatal depression, lethargy, apnea and hypothermia, especially at higher doses. Low dose acepromazine may be an option for very stressed or anxious dogs to avoid decreased uterine blood flow. The drug crosses the placenta slowly due to its higher molecular weight and protein binding, and is not associated with increased maternal or neonatal mortality 2 4, but its α-adrenergic antagonism can cause vasodilation and therefore it should be avoided in dehydrated or compromised patients.
  
  

• Opioids include mu-receptor agonists such as morphine, hydromorphone, oxymorphone, fentanyl, methadone and meperidine. Buprenorphine is a partial muagonist and butorphanol is a mu-receptor antagonist and a kappa agonist. These latter two drugs typically produce less sedation and respiratory depression than full mu agonists, but offer less potent analgesia. 
  Administering opioids to the dam will result in placental transfer, the extent of transfer differing depending on the drug. Less than 10% of buprenorphine is transferred to the fetus, whereas fentanyl (which is highly lipid soluble) crosses the placenta in high amounts, persisting long after maternal clearance 4. Of the commonly used mu agonists, morphine is the least lipid soluble and only 20-30% is non-ionized at normal plasma pH 4. It crosses the placenta less rapidly than more lipid soluble agonists such as fentanyl. The lipid solubility of hydromorphone lies between these two drugs. Neonates are significantly more sensitive to CNS and respiratory depression of opioids due to the immaturity of their CNS, increased permeability of the blood-brain barrier and end-organ sensitivity, and even small changes in ventilation may lead to hypoxemia and increased mortality. Respiratory and CNS depression in newly delivered neonates can be reversed with naloxone
  .

Anesthetic induction

The use of thiopental, ketamine, xylazine and methoxyflurane have been associated with increased puppy mortality and/or decreased puppy vigor at birth, and are best avoided 1 11 12. Epidural anesthesia/analgesia used alone for CS has minimal fetal effects but has various drawbacks; it is technically difficult, airway protection via intubation is impossible, and there is the risk of hind-limb paralysis, hypotension or urinary retention (if epidural opioids are used).

Propofol has a short half-life and rapid metabolism, including extra-hepatic metabolism, but it can cause cardiopulmonary depression depending on the dose and rate of administration. Propofol followed by isoflurane has puppy survival rates equivalent to epidural anesthesia and is associated with a positive effect on neonatal survival at 7 days 1. Propofol has similar fetal mortality rates to mask induction with isoflurane, but it allows IV induction and rapid control and protection of the airway 1 13.

Alphaxalone is available in many countries and has been shown to have a shorter terminal half-life compared to propofol 14 15. Two recent studies comparing propofol to alphaxalone for induction at CS found no significant difference in puppy mortality at 24 hours or up to 3 months after delivery, but both studies identified differences in “puppy vigor”. Apgar scores and all four health vigor assessments (withdrawal, sucking, anogenital and flexion reflexes) were greater for puppies for up to 60 minutes post-delivery where the dams received alphaxalone 16 17. 

Anesthetic technique

As noted above, the increased risk for regurgitation and aspiration can be decreased using maropitant (1.0 mg/kg SC administered at least 30 minutes prior to opioid administration 18). For more emergent situations, maropitant may be given (1.0 mg/kg IV over 5 minutes while monitoring blood pressure) and discomfort on injection can be decreased by diluting 50:50 with crystalloid solution. Alternatively, an opioid premedication which does not induce vomiting may be administered (e.g., butorphanol at 0.2-0.3 mg/kg IM or IV). Sedatives, e.g., acepromazine at 2.0-5.0 µg/kg IM or IV, should be reserved for highly stressed dams; however, remember that this drug has a long duration of action and cannot be reversed in either the dam or puppies.

Lumbosacral epidural with either local anesthetic or opioid (alone or in combination) can be administered either before or directly after induction to provide anesthesia and analgesia. Lidocaine (2.0-3.0 mg/kg, 0.1-0.15 mL/kg of 20 mg/mL concentration) will last for ~90 minutes, whilst bupivacaine (0.75-1.5 mg/kg, 0.1-0.2 mL/kg of 7.5 mg/mL concentration) lasts up to 4-6 hours. Both drugs will affect hindlimb motor function and may contribute to intra-operative hypotension through sympathetic blockade. Alternatively, preservative-free morphine may be used alone to provide analgesia (0.1-0.2 mg/kg); the onset of action may take up to 60 minutes, but motor function is not affected. Epidural opioids may cause urinary retention and – since early hospital discharge is advantageous for healthy dam and puppies – owners must monitor urine output for 24 hours after discharge. The combination of lidocaine (2.0 mg/kg) and morphine (0.1 mg/kg) provides rapid onset of anesthesia along with synergistic and long-lasting analgesia. Epidural drug administration requires much lower doses than the same drugs given parenterally and therefore decreases systemic effects in the dam and puppies. Alternatively, the epidural can be performed after the pups are delivered and the incision is closed to provide post-operative analgesia. 

If local anesthetic epidural is not performed, a lidocaine block along the line of incision (at 2 mg/kg, diluted with sterile water if necessary to increase the volume) will provide intra-operative analgesia. Bupivacaine may be used (1.5-2.0 mg/kg) for an incisional block at the time of closure of the linea alba for longer post-operative analgesia. Lidocaine and bupivacaine mixtures have a decreased duration of action and are not recommended 19.

Excretion of lidocaine and bupivacaine and/or their metabolites does occur but there is minimal effect on human neonates 22 and milk concentrations of ropivacaine are lower than for other local anesthetics 23. Although species differences may occur, based on current evidence it seems that most commonly used analgesics may be safely administered to lactating dams without adversely affecting the neonates.

Neonate resuscitation